Proceedings of a Workshop—in Brief

Navigating Infodemics and Building Trust During Public Health Emergencies

Proceedings of a Workshop—in Brief

INTRODUCTION

At the request of the Centers for Disease Control and Prevention (CDC), the National Academies of Sciences, Engineering, and Medicine hosted a two-day public workshop on April 10-11, 2023 to examine the history of public health infodemics, the impact of infodemics on trust in the public health enterprise, and tools and practices used to address infodemics. The first day of the workshop focused on developing an understanding of the modern health information ecosystem and examining the impact of infodemics during a public health emergency (PHE), discussing the public health capacities and capabilities for infodemic management during a PHE and throughout the disaster management cycle, and exploring the roles, responsibilities, and partnerships among relevant organizations and agencies in managing infodemics during a PHE. The second day of the workshop focused on learning about interventions to prevent and mitigate the harmful effects of infodemics and identifying research gaps.¹ This Proceedings of a Workshop—in Brief highlights key points made by workshop participants during the presentations, panel discussions, and responses to audience questions.² At the outset of the workshop, Howard Koh, Harvard T.H. Chan School of Public Health, described the term “infodemic” as the rapid spread of large amounts of sometimes conflicting or inaccurate information that can impede the ability of individuals, communities, and authorities to protect health and effectively respond in a crisis. Even a deluge of accurate information can overwhelm the public. In times of emergency, there may also be situations in which people do not have access to the information they need.

In his opening remarks, Kevin Griffis, CDC, described the COVID-19 pandemic as spotlighting the substantial effect disinformation can have on public health during an emergency. Addressing disinformation at scale is a relatively new challenge that has prompted various developing approaches, an emerging research base, and the use of novel tools such as AI. As part of a coordinated effort across the U.S. government, CDC is establishing more robust systems, processes, research,

__________________

trials, and evaluation to manage disinformation. The agency’s approach involves creating observational and alert systems and tools, increasing monitoring capacity, responding to misinformation, and encouraging public trust.

UNDERSTANDING THE COMPLEXITY OF INFODEMICS IN PUBLIC HEALTH EMERGENCIES

Panelists discussed historical infodemics and current contextual factors that enabled the COVID-19 infodemic to spread to an unprecedented scale. In describing the history, Nancy Tomes, Stony Brook University, noted that infodemics predate the internet and social media technology that have exacerbated them. Infodemics occurred during the 1918–1919 influenza pandemic, the 1980s HIV/AIDS pandemic, and the 2003 severe acute respiratory syndrome outbreak. In the early stages of each of those outbreaks—when scientific uncertainty was at its height and treatments and vaccines were not yet available—evolving science inadvertently generated confusion before adequate evidence and consensus were officially established. Infodemics flourished within the expectation gaps between an idealized faith in science and the realities of limited scientific evidence that can be produced at short notice. Her historical analysis of these events aligns with contemporary analyses in determining that information-related problems are not created by an abundance of information alone; the curation and filtering of information underpin infodemics. Furthermore, she noted, the social, economic, and political contexts of an epidemic are important to understanding and managing an infodemic (e.g., structural racism and misogyny, extreme income inequality, and distrust engendered by unresolved legacies of slavery and colonialism).

The speed and scale of infodemics now exceed the current capacities of health information technologies and underfunded health systems, said Tina Purnat, World Health Organization (WHO). Although infodemics have affected communities and health authorities in prior outbreak responses, the COVID-19 infodemic highlighted the speed and capacity of an increasingly digitized society’s ability to share information that simultaneously scales locally and globally. Purnat explained that over the past 15 years, narratives that influence knowledge, attitudes, perceptions, and behaviors related to public health emergencies have spread with increasing speed. To keep pace, health authorities must be quicker to identify and respond to circulating narratives. Moreover, the context of an emergency involves rapid and continual evolution of health information and evidence. Purnat highlighted the multidisciplinary expertise needed to manage infodemics and stated that effective infodemic management involves addressing behaviors and issues ranging from individual to structural and systemic. To this end, WHO trains infodemic managers in health informatics, epidemiology, behavioral science, health communication, digital analytics, and health promotion. In addition, WHO offers free online courses on infodemic management and has established a research agenda for managing infodemics.³

Dietram A. Scheufele, University of Wisconsin-Madison, described the effects of an amalgam of science, policy, and partisan politics on declining trust in science. During the COVID-19 pandemic, public health officials were asked to inform policy based on rapidly developing science under a heightened degree of public scrutiny. He noted the disruptive nature of interventions designed to mitigate COVID-19 transmission fueled partisan politics and policies—a dynamic that blurred the lines between expertise and advocacy. Current information ecologies can exacerbate bias, he added. These ecologies are based on delivery systems explicitly designed for commercial use and algorithmically curated to the preferences of the user. They feature AI intended to outperform human cognition, making it difficult to discern whether its output is artificially generated. Within this online environment, corrective measures to misinformation can increase its visibility. Additionally, inevitable variation in policy across different countries can cause confusion within the public. Scheufele explained that science-based decision making alone cannot eliminate variance or generate trust, as differing public health policy conclusions can be based on the same science.

__________________

Umair Shah, Washington State Department of Health, discussed rumors that affected behavior during the responses to Hurricanes Katrina in 2005 and Harvey in 2017 as well as rumors that fueled discrimination against Asian Americans during the COVID-19 pandemic. He distinguished misinformation—which can originate from people who do not know the information is inaccurate—from disinformation, which is incorrect information intentionally spread to cause chaos or political ramifications. Shah highlighted several factors that contribute to infodemics, including (1) the ease with which photos and videos can be manipulated; (2) the perpetual nature of the 24-hour news cycle; (3) the number and variety of media platforms; (4) media attention given to individuals lacking credibility; and (5) the global nature of online and social media platforms. Moreover, health emergencies generate collective anxiety that can lead to public denial of accurate information. He concluded that effective infodemic management requires diverse tools and capabilities that include “soft skills,” such as communications and engagement, in addition to proficiency in epidemiology and data collection.

EXPLORING INFODEMIC MANAGEMENT AND TRUST BUILDING APPROACHES

Tara Kirk Sell, Johns Hopkins Bloomberg School of Public Health, reviewed efforts to understand and respond to modern-day infodemics. She stressed that a constellation of interventions is needed to effectively meet the diverse needs of various communities and populations in addressing infodemics. Annie Sundelson, Johns Hopkins Bloomberg School of Public Health, coauthored the workshop’s commissioned paper with Sell and spoke about the approaches that can be used at the informational, individual, community, and institutional or structural levels of society to help manage infodemics as well as the evaluation of and evidence for these approaches.⁴ Information-level approaches include amplifying accurate information, filling information voids, debunking false information, tracking information, and utilizing tools to verify information, assess the credibility of sources, and detect false information. Evidence of success exists for many of these approaches and tools, but biases (e.g., confirmation bias) can substantially limit effectiveness. Individual-level resiliency approaches enhance information literacy or confer resistance to mis- or disinformation via pre-bunking or inoculation. Approaches at the interpersonal or community level include tools designed for public health communicators and community engagement efforts.⁵ Sundelson noted that leveraging community messengers and understanding information needs are long-term endeavors that offer promise for increasing trust in public health. Institutional or structural-level approaches include management of scientific literature; providing resources to infodemic managers, infodemic researchers, journalists, and fact checkers; social media regulation; and policy and legislation. Many of these approaches seek to address information issues before content reaches the public, and thus may be more effective than efforts to respond to misinformation that has already emerged. However, with regard to policy and legislation, she noted the tension between achieving public health aims and protecting civil liberties.

Sandra Quinn, University of Maryland, provided an overview of the Building Public Trust in Public Health Emergency Preparedness and Response (PHEPR) Science workshop held in 2022.⁶ During that workshop, participants spotlighted potential strategies for building and maintaining trust in PHEPR practice and science, particularly as it pertains to community engagement, the information environment, and public health infrastructure. Quinn emphasized the importance of humanity and relationship building with diverse communities, cross-sector collaboration, communication, capability, reliability, and transparency in building trust in the public health enterprise.

__________________

EXAMINING STRATEGIES AND INTERVENTIONS AT THE SYSTEMIC AND STRUCTURAL LEVELS

Workshop panelists discussed public health strategies and interventions to address infodemics at the systemic, structural, and institutional levels, exploring foundational concepts—terminology, principles, and frameworks—as well as practical approaches to infodemic surveillance, prevention, and response. Additional steps to prevent and respond to infodemics suggested by individual speakers are listed in Box 1.

Conducting Infodemic Surveillance

David Scales, Weill Cornell Medical College, discussed parallels between infectious disease surveillance and infodemic surveillance. Event-based surveillance for infectious disease relies on official and unofficial reports that yield early warnings for disease outbreaks, but the credibility of this approach is uncertain because of the unknown reliability of the reports. There is also no standard definition of “event,” which can provide some flexibility for actors doing such surveillance. For example, the CDC’s Vaccine Adverse Event Reporting System is an event-based reporting system with established limitations but is sometimes misinterpreted and tied to misinformation about COVID-19 vaccines. Indicator-based surveillance for infectious disease, however, depends on generally more reliable data sources (e.g., physician diagnosis, laboratory-confirmed diagnosis, or syndromic surveillance data) to monitor trends over time and inform decisions to a disease outbreak, and is typically associated with higher credibility. Scales suggested that infodemic surveillance often looks like event-based surveillance. Such an approach would provide actors doing infodemic surveillance flexibility to define what events should be tracked. Scales suggested that indicator-based surveillance for infodemics could be further evaluated and defined to focus on claims that are widely agreed to be false and of high potential for public health consequence.

The Pan American Health Organization/WHO Anti-Infodemic Virtual Center for the Americas (AIVCA) has established international, multidisciplinary teams to detect, research, prioritize, and respond to misinformation threats, explained Ian Brooks, University of Illinois Urbana-Champaign. Data scientists lead the detection team and utilize AI to monitor social media and follow other surveillance organizations for new and emerging misinformation. The evidence team, led by information scientists, identifies and curates information to aid senior public health officials in understanding new misinformation. Behavioral scientists lead the response team, crafting response narratives via a storytelling framework. AIVCA categorizes, evaluates, and prioritizes misinformation to direct response resources toward the narratives that are most likely to generate behavioral changes that could negatively affect public health.

Tricia Corrin, Public Health Agency of Canada (PHAC), described a COVID-19 literature surveillance framework created by PHAC to identify, characterize, and disseminate COVID-19 evidence daily. Initiated in January 2020, the framework used established knowledge synthesis methodologies and AI technology to launch a repository of COVID-19 literature tagged by category and key words. A team worked to fill information voids by sending daily emails to governments, local public health agencies, and academic institutions that highlighted timely and credible research. PHAC modeling teams utilized the data to create guidelines and website content. A collaboration with WHO enabled information extraction and reporting that avoided duplication and was useful to both agencies. After updating the framework for mpox, PHAC is now working to improve the literature surveillance system to employ more AI tools and include a dashboard to facilitate public database searches.

The NYC DOHMH Misinformation Unit was established in March 2021 to identify factors undermining COVID-19 vaccine confidence and has subsequently implemented misinformation surveillance, response, and evaluation strategies, said Maddie Perlman-Gabel, New York City Department of Health and Mental Hygiene (NYC DOHMH). In the culturally, ethnically, and racially diverse population of New York City, structural racism and community divestment have contributed to an unequal disease burden that disproportionately affects Black, Latino, immigrant, and low-income communities. The COVID-19 pandemic spotlighted these inequities, both in terms of disproportionately high hospitalizations

and low vaccine uptake. By analyzing data from industry reports and surveillance tools, social media, local news, and community feedback, the unit can conduct rapid community assessments of new misinformation. The Misinformation Unit learned that that public health infodemic management efforts should (1) prioritize equity and accessibility; (2) think globally and act locally; (3) grow external relationships; (4) foster internal collaboration; and (5) build capacity for monitoring and evaluation.

Preventing and Responding to Infodemics

Anne Zink, Alaska Department of Health, discussed approaches to sharing health information with the public. She remarked that people want to see the “me” within the “we” of data, in that if people do not see themselves represented, they can lose trust in information sources. “Public health information can be most effective when messaging seeks to connect, rather than to correct and connotes authority without being authoritarian,” stated Zink. Increasing public resistance to misinformation and receptiveness to accurate information involves building trust via accountability and communication, she explained. Effectively sharing public health information requires accounting for the foundational knowledge needed for the public to understand it, given the population’s continuum of education levels. She added that partnerships between government bodies, local public health departments, community-based organizations (CBOs), and trusted messengers within churches, school districts, and other bodies can effectively drive communication to the local level.

Sentiment analysis can be used to assess how the public perceives messaging in order to maximize its effectiveness, said Phil Maytubby, Oklahoma City County Health Department (OCCHD). Sentiment analysis is conducted across media types to identify words connected to messages that are perceived negatively by various demographics. After contracting sentiment analysis services, OCCHD changed wording within its COVID-19 vaccination campaign to increase positive public perception by emphasizing the word “choose” in their messaging, understanding that many Oklahomans perceive choice in the decision to be vaccinated as important. OCCHD also employed low-cost messaging promotion on social media to increase audience size.

Scott Burris, Temple University, discussed how public health law can be leveraged to address infodemics. “Law defines public health authority, can act as an intervention, and shapes and mediates social determinants of health,” he stated. Although legal processes provide opportunities to build and maintain trust, Burris noted that legally induced mistrust can fuel infodemics and hamper the effective use of law during public health emergencies. During the COVID-19 response, for example, public health missed opportunities to build trust in—and enhance compliance with—the rules that it issued. Burris and colleagues developed a legal module outlining essential public health law services and tasks required for effective legal action in public health (Burris et al., 2016). Burris elaborated that public health law services can help address infodemics by drafting laws better designed to avoid resistance and explaining and implementing the enacted rules more clearly. Much law-related work in the public health system is carried out by non-lawyers, who require additional training to be effective. Burris contended that the best way to build trust and compliance in health law is to design and execute it competently, which can be achieved by strengthening the public health workforce’s legal training and professional capacity.

Thomas Wilkinson, U.S. Department of Homeland Security, highlighted the role of community and trust building in federal infodemic response and asserted that a public health infodemic is a national security concern. Wilkinson explained that connections between local communities and federal government are tenuous and must be strengthened to effectively counter misinformation. Moreover, scientific hubris among health professionals can undermine their ability to effectively engage with patients, which can further ostracize already-marginalized communities.

COMMUNITY ENGAGEMENT FOR INFODEMIC PREPAREDNESS

Throughout the workshop, many panelists emphasized the importance of community engagement and trust building between crises in an effort to address health and information disparities, communicate health

information, and counter misinformation during an emergency.

Building Trust and Relationships with Communities

Relationship building and trust are the common denominators in successful crisis communication outcomes, according to Lisa Cox, Missouri Department of Health and Senior Services. “Effective emergency preparedness hinges on cultivating relationships with community leaders in advance of an event and communicating with them about public health goals,” said Cox. For example, public health officials leveraged strong, preexisting relationships with representatives from the Missouri Agriculture Department to reach farming communities. To engage with other hard-to-reach communities, they collaborated with trusted messengers ranging from religious leaders, physicians, officials from the Consulate of Mexico, and various respected and well-connected small-town residents. She added that pre-established relationships with media outlets— particularly those with political leanings at odds with public health aims—can increase the likelihood that those outlets will seek information from public health officials during a crisis.

Susan Kansagra, North Carolina Department of Health and Human Services (NCDHHS), also emphasized the importance of community-building and communication in public health initiatives. In addressing the COVID-19 infodemic, NCDHHS used an infrastructure of community-based partners to implement community outreach activities around vaccination. Over time, this initiative narrowed the vaccination disparities between Hispanic and Black populations compared to white individuals especially in older age groups. The NCDHHS communications team also held focus groups and conducted misinformation surveillance on social media and continually corrected messaging accordingly. For instance, they created short informational videos to counter the emerging false narrative that vaccination causes infertility. Kansagra suggested that public health could improve its public perception outside of crises with proactive messaging about the benefits of routine public health activities (e.g., protecting against food-borne illness by monitoring restaurants and protecting against lead exposure in and childcare facilities).

Infodemics seed distrust in public health, which makes it more difficult for individuals to access reliable information and guidance during an emergency, said Mysheika Roberts, Columbus Public Health. She explained that the speed and breadth with which social media can spread misinformation was illustrated not only during the COVID-19 pandemic but also during a 2022 measles outbreak in Columbus, Ohio that affected 85 unvaccinated children, 36 of whom required hospitalization. The majority were old enough to have been vaccinated, but their parents had refused based on the erroneous belief that vaccines cause autism. Roberts highlighted that an essential feature of an effective response to misinformation is transparency; this can be facilitated by sharing knowledge about a health threat through a publicly accessible dashboard featuring the number, demographics, and severity of cases. Such a dashboard can help counter misinformation when accurate information from the dashboard is disseminated by the media, social media influencers, community leaders, and other trusted messengers.

Infodemic responses should shift from a focus on communities “getting the facts wrong” to public health authorities “getting the right facts” by designing and delivering services that communities deem trustworthy, beneficent, and useful, said Monica Schoch-Spana, Johns Hopkins Bloomberg School of Public Health. Engaging with communities to seek their ideas and practical support for preparedness and response systems can yield public health interventions that are more culturally competent, attuned to local conditions, and socially acceptable. Community-based participatory research techniques can reveal nuanced values and priorities within vulnerable populations. For example, a community engagement team working with Hispanic farmers in rural Idaho learned that work was a primary value within that community and aligned COVID-19 interventions accordingly (e.g., highlighting how testing and vaccination can reduce work disruptions, holding vaccine clinics at local farms and food processing

facilities). Schoch-Spana underscored the value of funding, experienced staff, and formal community engagement policy in strengthening participatory approaches to public health emergency preparedness.

Teresa Wagner, University of North Texas Health Science Center at Fort Worth, highlighted support for health literacy as an approach to counteracting harm from infodemics. She explained that in recent years, the American public has received a high volume of health information from myriad sources that may not help individuals develop a solid understanding of complex information. Health literacy is a global resource and a community asset that decreases the harm caused by misinformation. Approximately 90 percent of U.S. adults have limited health literacy, which is more common among older adults, people of color, low-income individuals, and people without a high school diploma. As trusted community members, community health workers (CHWs) are well-positioned to deliver understandable health information to the individuals they serve. Just as CHWs assist people in navigating complicated health care systems, they can help individuals navigate misinformation during public health emergencies. Wagner highlighted that in urban and rural areas with little to no access to health care services, building CHW workforce capacity and upskilling to address complex medical topics can fill a critical gap in those communities that supports day-to-day wellness as well as bolstering preparation for future crises.

Supporting Marginalized Communities

Norman Oliver, formerly Virginia Department of Health (VDH), considered how to effectively address infodemics within marginalized communities. He emphasized that Black and Brown, rural, and low-income communities suffer disproportionately negative outcomes in public health emergencies—including infodemics—and thus responses should be structured to integrate the needs of these communities. For example, VDH Incident Command Structures assigns personnel to apply an equity lens in reviewing the organization’s communication, finances, logistics, operations, and planning. Oliver explained that efforts to counter misinformation could be tailored to the demographics, political beliefs, and values of target populations. Crafting this targeted messaging benefits from input from communication science experts and support from a communications workforce within public health. Additionally, identifying and enlisting trusted messengers (both individuals and organizations) within targeted communities can increase community receptiveness.

Eve Shapiro, West Side United, provided an overview of the Racial Equity Rapid Response Team (RERRT),⁷ a data-driven, community-based mitigation effort to reduce illness and death from COVID-19 in Black and Brown Chicago communities. West Side United, a health collaborative composed of six hospitals and a variety of community partners, established RERRT to address the racial disparities seen in Chicago’s COVID-19 mortality rates. RERRT provided residents with health information, supplies such as personal protective equipment, COVID-19 testing, and support services. The team featured working groups focused on (1) developing COVID-19 prevention messaging for priority audiences; (2) building responsive testing and treatment strategies; (3) direct services to community members impacted by COVID-19; (4) resources; (5) data; and (6) a steering committee to connect the efforts of all other working groups. For example, the data and direct services groups conducted a survey to understand community perspectives and beliefs about COVID-19 that then informed the work of the messaging group. Interdisciplinary collaboration and community dialogue were foundational to these mitigation efforts, said Shapiro.

Amanda Boyd, Washington State University, explored infodemic preparedness and response efforts within Indigenous communities. She explained that health communication efforts are most successful when they honor the experience and expertise already present in native communities—such as tribal elders and medical professionals—who understand local perceptions of and barriers to health programming. Engaging with these trusted messengers can help incorporate cultural and

__________________

traditional norms into public health communications; their role is especially valuable given the history of atrocities against Indigenous peoples and consequent distrust toward outside messengers, researchers, governments, and organizations. During the COVID-19 pandemic, some vaccine hesitancy within Indigenous communities stemmed from beliefs that traditional medicines were adequate prevention and concerns about how vaccines would affect Indigenous mRNA, underlining the need for culturally appropriate communication strategies. Jill Jim, National Indian Health Board, remarked that many Indigenous communities are in rural and remote areas lacking broadband access, so it is important to select communication channels strategically. For instance, in many tribal nations and reservations, radio is more reliable than internet and local newspapers are popular. Some Indigenous people with online access subscribe to digital news sources such as Indian Country Today or follow news stories on social media. Many tribal health facilities create video content to play in waiting rooms.

The COVID-19 infodemic had harmful consequences for many Asian, Asian American, and Pacific Islander (AAPI) populations, remarked Juliet Choi, Asian & Pacific Islander American Health Forum (APIAHF). For example, hate language associating the COVID-19 virus with China fueled acts of violence that forced AAPI communities to contend simultaneously with hate and the ramifications of the pandemic and infodemic. APIAHF created toolkits and messaging to support advocacy efforts. The “model minority” myth—which purports that AAPI individuals are more successful than the general population—can result in reduced allocations of resources, thereby limiting assistance during emergencies. She also noted that trusted messengers are critical for providing targeted, language-accessible information to AAPI communities. More than 40 percent of adults in Chinese, Bangladeshi, Vietnamese, Nepalese, and Burmese communities in the United States have limited English skills that hinder their understanding of health information. During the COVID-19 pandemic, APIAHF worked to address the language barrier by hosting online video meetings with simultaneous interpretation.

Claudia Regalado-Cattouse, Chicago Cook Workforce Partnership, gave an overview of the Chicago Community Health Response Corps, a community outreach effort created in June 2020 to conduct COVID-19 contact tracing and vaccination education.⁸ This economic development and public health initiative utilized grant funds to generate employment opportunities during the pandemic, connected the public to government agencies, and fostered trust building. Community hires contacted their fellow community members by phone and door-to-door visits. To address the resistance many individuals initially expressed about sharing personal information, workers strived to understand their needs, answer their questions, and connect them to resources. Dorothy Oliver, community advocate for Panola, Alabama, described a similar approach to effective outreach in her community. She recounted efforts to maximize COVID-19 vaccination uptake in Panola, which involved contacting community members by phone and door-to-door visits, collaborating with local leaders, visiting churches, and talking with customers in her convenience store. Noting that emotions run high during health emergencies, she emphasized the importance of calmly communicating accurate and culturally relevant information.

INNOVATIVE INTERVENTIONS AND COMMUNICATION STRATEGIES

Panelists discussed how information ecosystems can contribute to infodemics and highlighted interventions to address misinformation, both within these systems and at the individual level. Panelists explored strategies to better understand and support the public in infodemic preparedness and response, with a particular focus on how to support diverse communities. They explored strategies to map and monitor online ecosystems as well as innovative interventions and communication strategies to counter misinformation.

Mapping and Monitoring Online Ecosystems

Mapping the ecosystem of online communities can predict the potential spread of misinformation, said Neil Johnson, George Washington University. Johnson discussed how small but active online communities

__________________

can negatively affect state and national interventions; mapping can anticipate this resistance and provide triggers for the public health response to misinformation. For instance, researchers can identify multiple online communities with a shared element—for example, many parenting, autism, and alternative health groups share vaccine distrust—and then determine the likelihood that they will share information with one another, revealing connections between local and global facets of the shared issue.

Alejandro Posada, Internews, provided an overview of Rooted in Trust (RiT),⁹ a global pandemic information response program that conducts community-driven rumor tracking and social listening, analysis, and global level research on infodemic management. Social listening is a practical tool that synthesizes public discourse into insights about a particular community; these insights can inform efforts to establish and strengthen connections between stakeholders. The resulting networks coordinate efforts to improve the accuracy and appropriate timing of communications. With the goal of promoting resilient and better-prepared information ecosystems for marginalized and at-risk communities, RiT works in 31 languages and produces rumor bulletins and radio broadcasts. Its research arm is currently testing the hypothesis that effective infodemic preparedness within vulnerable communities involves building and maintaining strong community ecosystems of civil society organizations and local leaders. These ecosystems contextualize science-based information from academia and health agencies and facilitate bidirectional communication with communities.

Christopher Voegeli, CDC, highlighted the State of Vaccine Confidence Insights Report series,¹⁰ in which CDC provides iterative and inductive analyses of misinformation, tracks themes, and issues recommendations. Data sources for this series range from Google trends to CDC to third-party reports. Once multiple inputs are synthesized and themes are identified, CDC uses a threat-classification matrix to estimate the scale of misinformation spread and the populations it is likely to affect. These quarterly reports, which focus on specific topics or demographic/geographic populations and recommend actions, are currently received by more than 1,000 CBOs and government organizations. According to a survey, most readers utilize the series to inform communication strategies and to better understand vaccine hesitancy across special populations. Voegeli contended that debunking misinformation will have limited effectiveness unless the larger social and cultural forces that bolster misinformation are addressed, which will require evidence-based approaches that leverage socio-behavioral and epidemiological insights.

Data from information ecosystems can be employed by private interests to influence bipartisan issues, cautioned Richard Legault, U.S. Department of Homeland Security (DHS). Thus, both data usage and the ways the public consumes information can have ramifications for public health and national security, warranting further research to predict how information ecosystems will operate in the near- and long-term. Legault emphasized that research is also needed to systematically measure the effects of communications on public health outcomes. It is challenging to measure outcomes due to the rapid rate of change within information ecosystems, the difficulty in connecting online behavior and information consumption, and resistance to infodemics interventions. Developing the ability to understand drivers of behavior and quickly adapt to changes within information ecosystems is a medical-socio-technical endeavor, he noted.

Interventions at the Informational and Individual Levels

Psychological inoculation, or “pre-bunking,” is a communication technique that builds psychological resistance against a future persuasion strategy, explained Rachael Piltch-Loeb, Harvard T.H. Chan School of Public Health. Pre-bunking involves (1) warning people that they may encounter false or misleading information; (2) offering counterarguments to misinformation to equip people to refute manipulative messaging; and (3) providing a micro-dose of the misinformation that people may encounter. This three-part strategy can

__________________

be used in text-based and video-based interventions. Inoculation-based messaging has demonstrated potential in increasing an individual’s ability to identify misinformation and decrease the likelihood that they share misinformation.

Jessica Castner, University of Albany, suggested that nurses are uniquely positioned to mitigate the harmful effects of infodemics, both because the profession is among the most trusted in the United States and because nurses are more likely to practice in rural and low-income regions of the country than other care providers. Moreover, nurses practice in settings that enable frequent interactions with the public, such as schools, incarceration settings, and community programs. She suggested that interventions to provide health information would benefit from focusing on active listening to better understand an individual’s values and barriers; this would enable more meaningful communication. She recommended that health information should be provided (1) within a person’s environment; (2) in a format that is sensitive to culture, feelings, literacy, language, priorities, and core beliefs; (3) by/with a trusted source; (4) in a manner that highlights benefits or aligned values and beliefs; and (5) repeatedly.

Joe Smyser, The Public Good Projects (PGP), remarked that infodemiology is a relatively new field, offering a valuable opportunity for new stakeholders and partners to approach public health in creative ways that build trust within the public; this is coupled with the opportunity to rebuild a stronger public health system after the COVID-19 pandemic. He maintained that infodemic management teams should be the foundation of health communications from federal, state, and local health agencies moving forward. Health communications should move beyond crisis communications responses to relationship building with stakeholders to operationalize public health concepts. He provided an overview of the PGP vaccine misinformation program, which operates with advanced media monitoring systems to track, tag, and respond to a range of priority public health topics.¹¹

Adam Dunn, University of Sydney, discussed information exposure research utilizing new data collection technology. His initial data-driven research on the effects of health misinformation on behaviors demonstrates that it is possible to accurately predict—based solely on the accounts they follow—whether a Twitter user’s first tweet about human papillomavirus will be in support of or against vaccination (Dunn et al., 2015). He noted that current social listening and data-driven research methods account for misinformation on social and general media, but they are not designed to robustly measure the content that people are actually exposed to nor the effects of this information on attitudes or behaviors. This type of research could guide the prioritization of interventions and resource allocation toward vulnerable populations most at risk of harm and most susceptible to misinformation. In partnership with WHO, Dunn and colleagues have worked to address this gap by creating the Information Diary Platform (IDP), a smartphone-based application for logging encounters with information about health problems in contexts including social media, websites, newspapers, television, advertisements, clinic posters, and conversations with health professionals, loved ones, or acquaintances. In a six-click process, participants record the encounter and the level of trust they have in the information. The granularity of IDP data enables local public health interventions to target specific subpopulations. This data analysis can also be used to identify common risk signals across locations and populations to inform the priority of interventions.

Evaluating Rumors and Credible Sources

Diminished public trust in officials, coupled with increasing diversity of online platforms, creates a context vulnerable to the spread of misinformation—both accidental and intentional—that is exacerbated during crises, said Kate Starbird, University of Washington. She explored the benefits of shifting the terminology around infodemics. For instance, if the term “rumor” is defined as unverified or unofficial information transmitted through informal networks, it does not carry the stigma of being necessarily incorrect. Thus, use of the term “rumor” avoids premature labeling of information as false and can help crisis communicators build trust by acknowledging their audience’s agency. Starbird and

__________________

colleagues developed a 10-part framework to assess the probable spread of a rumor and elucidate how an emerging event could drive rumoring; it encompasses information conditions, contextual factors, and systems effects surrounding a rumor (Zeng et al., 2016). Starbird concluded that this framework can aid information managers in determining whether to respond to a rumor, given that addressing it can increase its visibility.

Sue Curry, University of Iowa College of Public Health, provided an overview of work completed by two advisory committees to define principles and attributes of credible health information sources. For online sources of health information, the foundational principles are that a credible source is science-based, objective, transparent, accountable, inclusive, and equitable. She noted that identifying credible sources are important but insufficient in addressing infodemics; social media platforms need to advance content assessment, misinformation management, health literacy, cultural competence, and research and evaluation.

ADDRESSING FUTURE INFODEMICS

During the concluding roundtable discussion, panelists identified practices, capacities, capabilities, partnerships, and collaborations that could be developed within the next decade to address infodemics. Strategies to communicate most effectively should be embedded into every aspect and at every level of public health, not just after decisions are made, maintained William Hallman, Rutgers Global Health Institute. He continued that this would require bringing on board professional communications experts with the requisite investment in training and the establishment of communications positions within public health. During crises, communications professionals can support the provision of rapid communication under conditions of uncertainty with sufficient transparency that the current understanding of a health emergency may change as the situation evolves. He cautioned that bureaucratic processes developed in response to legal and liability concerns can impede such communications, underscoring the need for expertise. Demetre Daskalakis, White House National Mpox Response, added that to avoid repeating past missteps, public health and academia should train local officials to understand and apply risk communication strategies in an effort to avoid absolute statements that often generate mistrust and lack of confidence. Hallman suggested creating the appropriate structures to (1) facilitate communication between public health professionals; (2) collect public perspectives, concerns, and questions on a routine basis; and (3) evaluate, record, and share the effectiveness of various interventions.

Michael Fraser, Association of State and Territorial Health Officials, also asserted that public health agencies should improve practices for obtaining and incorporating community feedback and input, as community engagement efforts should be at the forefront of infodemic response strategies. Hallman recommended shifting from top-down community engagement strategies toward structures to facilitate bottom-up communications from communities to public health

organizations. Structures for collecting community concerns could involve health care providers, professional medical associations, social media platforms, local media sources, businesses, and marketers of relevant products.

Sustaining robust emergency communications strategies between crises can increase the speed of infodemic response, said Daskalakis. Public health has an opportunity to avert a future panic-neglect response cycle by maintaining relationships—with trusted messengers as well as trusted platforms—and strengthening communications infrastructure that were developed in response to the COVID-19 pandemic and other past crises. For example, the communications infrastructure created and maintained to address HIV was leveraged in the response to the 2022 mpox outbreak. The mpox response then increased messaging visibility by partnering with online dating applications, a channel that can be leveraged to bolster future responses.

Public health should create engaging, entertaining content to generate and sustain focus on health issues between crises, suggested Bruce Y. Lee, City University of New York. Public health should draw upon the approach to communication used by private-sector companies, which integrate public perceptions and marketing into all phases of product development. Fraser suggested that the lackluster websites for most public health agencies are missed opportunities to connect with the public. Hallman remarked that messaging should be better targeted to key audiences and monitored to track its effectiveness. He added that much of the misinformation during the COVID-19 infodemic pertained to the presumed intentions of leaders and institutions rather than on claims of incorrect data, demonstrating that scientific facts are insufficient for addressing the emotional component of infodemics. To refine and focus communications, Daskalakis suggested that designing efforts to highlight the perspectives of Black, Latinx, Indigenous, and LGBTQIA+ populations could create connectivity and add nuance to infodemic responses. Additionally, because the public experienced trauma during the COVID-19 pandemic, a trauma-informed care lens is appropriate for approaching public health.

Creating a community of practice around addressing infodemics can leverage interdisciplinary expertise, suggested Fraser. Jewel Mullen, University of Texas at Austin, added that such communities of practice should strive for inclusion to avoid further marginalizing disenfranchised populations. Daskalakis highlighted the opportunity for public health to creatively partner with industry for mutual benefit, such as home-based testing companies developing products for sexually transmitted infections. Daskalakis commented that cross-sector relationship building can elevate the value of health within society. Fraser contended that infodemic preparedness should involve continued employment for CHWs as part of the community of practice. Many CHWs are currently being laid off as the official declaration of the end of the COVID-19 public health emergency is approaching, despite a host of opportunities for CHWs to maintain relationships within communities outside of public health crises (e.g., addressing chronic conditions). Mullen remarked that sustainable professionalization of CHWs would support a framework in which community engagement plays a role in detecting infodemics.

Lee recommended that public health should anticipate and leverage emerging technological innovation to shift from a reactive to a proactive response to infodemics. Additionally, including technology courses in public health degrees could improve basic technology proficiency. As public health employs new technology, Daskalakis emphasized the importance of transparency in data usage to maintain community trust. Hallman suggested exploring the possibility of building and training new technology platforms to prioritize responsible and authoritative information.

CONCLUDING REMARKS

In his closing remarks, Griffis called for increased understanding of how to resource institutions at all levels to reduce the harmful effects of mis- and disinformation. To this end, CDC conducts regular meetings with other federal agencies to coordinate responses to emerging misinformation. Furthermore, CDC is working to: (1) build improved misinformation monitoring and alert systems; (2) develop a more systematic approach to

misinformation through the agency; (3) create, in collaboration with academic institutions, a system to increase sentinel data collection and social listening; (4) establish rapid response infrastructure to provide the public with accurate information from trusted community sources; and (5) develop tools for public health to predict the virality of vaccine misinformation. CDC continually strives to disseminate accurate information as it is needed through appropriate channels in order to empower the public to make decisions that support their health and wellbeing.

REFERENCES

Burris, S., M. Ashe, D. Blanke, J. Ibrahim, D. E. Levin, G. Matthews, M. Penn, and M. Katz. 2016. Better health faster: The 5 essential public health law services. Public Health Reports 131(6):747-753.

Dunn, A. G., J. Leask, X. Zhou, K. D. Mandl, and E. Coiera. 2015. Associations between exposure to and expression of negative opinions about human papillomavirus vaccines on social media: an observational study. Journal of Medical Internet Research 17(6):e144. https://doi.org/10.2196/jmir.4343.

Zeng, L., K. Starbird, and E. S. Spiro. 2016. Rumors at the speed of light? Modeling the rate of rumor transmission during crisis. 49th Hawaii International Conference on System Sciences. Koloa, HI. Pp. 1969-1978.

DISCLAIMER This Proceedings of a Workshop—in Brief has been prepared by Anna Nicholson and Tamara Haag as a factual summary of what occurred at the meeting. The statements made are those of the rapporteurs or individual workshop participants and do not necessarily represent the views of all workshop participants; the planning committee; or the National Academies of Sciences, Engineering, and Medicine.

*The National Academies of Sciences, Engineering, and Medicine’s planning committees are solely responsible for organizing the workshop, identifying topics, and choosing speakers. The responsibility for the published Proceedings of a Workshop—in Brief rests with the institution. The planning committee comprises Howard K. Koh, Harvard T.H. Chan School of Public Health and the Harvard Kennedy School; Allison Arwady, Chicago Department of Public Health; Jessica Baggett, Association of State and Territorial Health Officials; Timothy Collins, Alaska Native Tribal Health Consortium; Lee A. Fleisher, Centers for Medicare and Medicaid Services; Kathleen H. Jamieson, University of Pennsylvania; Hilary N. Karasz, Public Health – Seattle & King County; Bruce Y. Lee, City University of New York; Gretchen Michael, Administration for Strategic Preparedness and Response; Jewel Mullen, University of Texas at Austin; Elena Savoia, Harvard T.H. Chan School of Public Health; Tara Kirk Sell, Johns Hopkins Center for Health Security; Jennifer Shopkorn, U.S. Census Bureau; Kasisomayajula Viswanath, Harvard T. H. Chan School of Public Health; Elisabeth Wilhelm, Brown University; and Seema Yasmin, Stanford University.

REVIEWERS To ensure that it meets institutional standards for quality and objectivity, this Proceedings of a Workshop—in Brief was reviewed by Phil Maytubby, Oklahoma City County Health Department; Rachael Piltch-Loeb, Harvard T.H. Chan School of Public Health; and Teresa Wagner, The University of North Texas Health Science Center at Fort Worth. Leslie Sim, National Academies of Sciences, Engineering, and Medicine served as the review coordinator. We also thank staff member Lida Beninson for reading and providing helpful comments on this manuscript.

STAFF Lisa Brown, Matthew Masiello, Margaret McCarthy, Board on Health Sciences Policy, Health and Medicine Division, National Academies of Sciences, Engineering, and Medicine.

SPONSORS This workshop was supported by the Centers for Disease Control and Prevention.

For additional information regarding the workshop, visit https://www.nationalacademies.org/event/04-10-2023/navigating-infodemics-and-building-trust-during-public-health-emergencies-a-workshop

SUGGESTED CITATION National Academies of Sciences, Engineering, and Medicine. 2023. Navigating infodemics and building trust during public health emergencies: Proceedings of a workshop–in brief. Washington, DC: The National Academies Press. https://doi.org/10.17226/27188.

Health and Medicine Division Copyright 2023 by the National Academy of Sciences. All rights reserved.